Instructions to use zai-org/GLM-5.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zai-org/GLM-5.2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="zai-org/GLM-5.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zai-org/GLM-5.2")
model = AutoModelForCausalLM.from_pretrained("zai-org/GLM-5.2", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zai-org/GLM-5.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zai-org/GLM-5.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zai-org/GLM-5.2

SGLang

How to use zai-org/GLM-5.2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "zai-org/GLM-5.2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "zai-org/GLM-5.2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zai-org/GLM-5.2 with Docker Model Runner:
```
docker model run hf.co/zai-org/GLM-5.2
```

sunyiyou commited on 6 days ago

Commit

4e53bee

verified ·

1 Parent(s): cf457fa

Add official Agents Last Exam evaluation results

Browse files

Adds ALE results from the [official leaderboard](https://agents-last-exam.org/leaderboard), captured on 2026-10-01.

- Covers Overall, ALE-CLI, Near-term, Full-spectrum, and Last-exam, with pass rate and score on a 0-100 scale.
- Configuration: GLM-5.2, Claude Code; effort=max. Task coverage is recorded in `notes`.
- Overall: 20.39% pass rate / 41.09% score.

These are existing official leaderboard aggregates, not newly rerun or HF-verified evaluations. The full benchmark contains 152 tasks, including unattempted tasks in the denominator.

Files changed (1) hide show

.eval_results/ale-claude-code-max.yaml +120 -0

.eval_results/ale-claude-code-max.yaml ADDED Viewed

	@@ -0,0 +1,120 @@

+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: overall-pass-rate
+  value: 20.394736842105264
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=146/152; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: overall-score
+  value: 41.08540680857251
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=146/152; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: ale-cli-pass-rate
+  value: 23.809523809523807
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=104/105; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: ale-cli-score
+  value: 44.15980756094993
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=104/105; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: near-term-pass-rate
+  value: 32.83582089552239
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=66/67; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: near-term-score
+  value: 60.33126287817978
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=66/67; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: full-spectrum-pass-rate
+  value: 18.181818181818183
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=54/55; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: full-spectrum-score
+  value: 36.974012794692094
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=54/55; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: last-exam-pass-rate
+  value: 2.631578947368421
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=34/38; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.
+- dataset:
+    id: agents-last-exam/agents-last-exam
+    task_id: last-exam-score
+  value: 12.862606139319707
+  source:
+    url: https://agents-last-exam.org/leaderboard
+    name: ALE official leaderboard
+    user: agents-last-exam
+  notes: harness=claude_code; harness_variant=max; attempted=34/38; official leaderboard
+    snapshot=2026-10-01T20:39:17.826877+00:00; values are percentages (0-100); equal-task
+    aggregation; unattempted tasks contribute zero; pass threshold score>=0.999. This
+    is an official leaderboard export, not an HF-verified evaluation.